From wolfgang.yang at gmail.com Thu Jun 3 12:35:20 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Thu, 3 Jun 2010 11:35:20 -0500 Subject: [Kim-discussion] Fwd: [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> Message-ID: Dear List, This thread is wrongly posted in interested-in-kim maillist, and I forward it here. Fangkai ---------- Forwarded message ---------- From: Yang Fangkai Date: 2010/6/3 Subject: Re: [Interested-in-kim] problem of populating instances from my own corpus To: Anton Andreev Cc: interested-in-kim at ontotext.com Anton, On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev wrote: > Hello Fangkai, > > First I would like to point out that the kim-discussion: > http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for asking > technical questions like this one. Next time please use the kim-discussion > mailing list, not this one. Thanks. > Sorry for the mistake. I will use that list the next time. > Now back to your problem: > What version of KIM do you use? KIM 2.4? > Yes. I am using KIM2.4 under Windows XP. > Are you using the KIMGate hybrid - a GATE developer with KIM's default > pipeline or the tool called "populater" again from the bin folder? I started KIM by running startkim.bat, and the populator by running toolPopulate.cmd in tool folder. I didn't see the tool "populator" in the bin folder. >The later > only needs a document source folder and uses an already running KIM > instance. Do you see that the documents are being annotated? What results do > you expect, what is missing? > Here is what I expect. I have a corpus containing about 2000 docs, and I want to query over these docs. So I plan to use toolPopulate to extract entities over these docs (this is what I am trying to do), and then query over them. I expect to see the entities populated from these docs, but I didn't see any meaningful entities when I query the entity from the KIM GUI. I don't know if the above makes sense. Thank you! Fangkai > The steps you are doing are correct in general. > > Best regards, > Anton Andreev > > -- > Anton Andreev > Account Manager > Ontotext AD > Tel: +359 2 875 81 17 > Fax:+359 2 975 32 26 > email: anton.andreev at ontotext.com > www.ontotext.com > > > > On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >> >> Dear List, >> >> ? ? ? ? ?I am trying to use Populate GUI to populate entities from my >> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >> articles from Wall Street Journal in plain text form, and refer to the >> folder in Populate GUI. However, it seems no entities is populated. I >> try to add an .xml file with the same name of the text file, but still >> doesn't work. (I check that by first deleting all files from >> /context/default/populated, and populate entities from a file, and >> check the entities by querying the entities at >> http://localhost:8080/kim, but no meaningful entities found). I am >> wondering if I miss some steps or important configurations. Thank you >> very much! >> >> Best, >> >> Fangkai >> _______________________________________________ >> interested-in-KIM mailing list >> interested-in-KIM at ontotext.com >> http://ontotext.com/mailman/listinfo/interested-in-kim >> > > > From wolfgang.yang at gmail.com Thu Jun 3 13:00:40 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Thu, 3 Jun 2010 12:00:40 -0500 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> Message-ID: hi, Anton, I tried HTML files, and the population works. But this just doesn't work for txt file... I checked the populator.xml and found the following configuration: doc,htm,html,txt,page,xml I suspect the populator has already been configured to process txt file. So where is the problem? Thank you! Fangkai 2010/6/3 Yang Fangkai : > Anton, > > On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev > wrote: >> Hello Fangkai, >> >> First I would like to point out that the kim-discussion: >> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for asking >> technical questions like this one. Next time please use the kim-discussion >> mailing list, not this one. Thanks. >> > > Sorry for the mistake. I will use that list the next time. > >> Now back to your problem: >> What version of KIM do you use? KIM 2.4? >> > > Yes. I am using KIM2.4 under Windows XP. > >> Are you using the KIMGate hybrid - a GATE developer with KIM's default >> pipeline or the tool called "populater" again from the bin folder? > > I started KIM by running startkim.bat, and the populator by running > toolPopulate.cmd in tool folder. I didn't see the tool "populator" in > the bin folder. > >>The later >> only needs a document source folder and uses an already running KIM >> instance. Do you see that the documents are being annotated? What results do >> you expect, what is missing? >> > > Here is what I expect. I have a corpus containing about 2000 docs, and > I want to query over these docs. So I plan to use toolPopulate to > extract entities over these docs (this is what I am trying to do), and > then query over them. I expect to see the entities populated from > these docs, but I didn't see any meaningful entities when I query the > entity from the KIM GUI. > > I don't know if the above makes sense. Thank you! > > Fangkai > > >> The steps you are doing are correct in general. >> >> Best regards, >> Anton Andreev >> >> -- >> Anton Andreev >> Account Manager >> Ontotext AD >> Tel: +359 2 875 81 17 >> Fax:+359 2 975 32 26 >> email: anton.andreev at ontotext.com >> www.ontotext.com >> >> >> >> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >>> >>> Dear List, >>> >>> ? ? ? ? ?I am trying to use Populate GUI to populate entities from my >>> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >>> articles from Wall Street Journal in plain text form, and refer to the >>> folder in Populate GUI. However, it seems no entities is populated. I >>> try to add an .xml file with the same name of the text file, but still >>> doesn't work. (I check that by first deleting all files from >>> /context/default/populated, and populate entities from a file, and >>> check the entities by querying the entities at >>> http://localhost:8080/kim, but no meaningful entities found). I am >>> wondering if I miss some steps or important configurations. Thank you >>> very much! >>> >>> Best, >>> >>> Fangkai >>> _______________________________________________ >>> interested-in-KIM mailing list >>> interested-in-KIM at ontotext.com >>> http://ontotext.com/mailman/listinfo/interested-in-kim >>> >> >> >> > > > > -- > Fangkai Yang, Ph.D student > Taylor Hall 3.150A > Department of Computer Sciences > The University of Texas at Austin > Austin, 78712-0233, Texas > USA > http://www.cs.utexas.edu/~fkyang > email: fkyang at cs.utexas.edu > -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu From Anton.Andreev at ontotext.com Fri Jun 4 06:45:36 2010 From: Anton.Andreev at ontotext.com (Anton Andreev) Date: Fri, 04 Jun 2010 13:45:36 +0300 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> Message-ID: <4C08D950.1070504@ontotext.com> Hello Fangkai, 1. It should work the way you are doing it. KIM 2.4 can be feeded with txt files. When you start "toolPopulate.cmd" do you see that documents are being processed? Also you can check the KIM server output. There should be some output which will prove that documents are being annotated. 2. Please go to the "Keyword Search" menu and try searching there for some of your content. Please keep in mind that: * CORE Search * Semantic Search * Timelines require you to configure CoreDB with Oracle: http://www.ontotext.com/kim/doc/sys-doc/ConfigCORE.html We know that the Oracle dependency is uncomfortable, so the next version of KIM 3.0 does not require Oracle. KIM 3.0 will be released in two weeks (if no major issues are found). Best regards, Anton Andreev -- Anton Andreev Account Manager Ontotext AD Tel: +359 2 875 81 17 Fax:+359 2 975 32 26 email: anton.andreev at ontotext.com www.ontotext.com On 3.6.2010 ?. 20:00 ?., Yang Fangkai wrote: > hi, Anton, > > I tried HTML files, and the population works. But this just > doesn't work for txt file... > > I checked the populator.xml and found the following configuration: > > doc,htm,html,txt,page,xml > > I suspect the populator has already been configured to process > txt file. So where is the problem? Thank you! > > Fangkai > > 2010/6/3 Yang Fangkai: > >> Anton, >> >> On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev >> wrote: >> >>> Hello Fangkai, >>> >>> First I would like to point out that the kim-discussion: >>> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for asking >>> technical questions like this one. Next time please use the kim-discussion >>> mailing list, not this one. Thanks. >>> >>> >> Sorry for the mistake. I will use that list the next time. >> >> >>> Now back to your problem: >>> What version of KIM do you use? KIM 2.4? >>> >>> >> Yes. I am using KIM2.4 under Windows XP. >> >> >>> Are you using the KIMGate hybrid - a GATE developer with KIM's default >>> pipeline or the tool called "populater" again from the bin folder? >>> >> I started KIM by running startkim.bat, and the populator by running >> toolPopulate.cmd in tool folder. I didn't see the tool "populator" in >> the bin folder. >> >> >>> The later >>> only needs a document source folder and uses an already running KIM >>> instance. Do you see that the documents are being annotated? What results do >>> you expect, what is missing? >>> >>> >> Here is what I expect. I have a corpus containing about 2000 docs, and >> I want to query over these docs. So I plan to use toolPopulate to >> extract entities over these docs (this is what I am trying to do), and >> then query over them. I expect to see the entities populated from >> these docs, but I didn't see any meaningful entities when I query the >> entity from the KIM GUI. >> >> I don't know if the above makes sense. Thank you! >> >> Fangkai >> >> >> >>> The steps you are doing are correct in general. >>> >>> Best regards, >>> Anton Andreev >>> >>> -- >>> Anton Andreev >>> Account Manager >>> Ontotext AD >>> Tel: +359 2 875 81 17 >>> Fax:+359 2 975 32 26 >>> email: anton.andreev at ontotext.com >>> www.ontotext.com >>> >>> >>> >>> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >>> >>>> Dear List, >>>> >>>> I am trying to use Populate GUI to populate entities from my >>>> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >>>> articles from Wall Street Journal in plain text form, and refer to the >>>> folder in Populate GUI. However, it seems no entities is populated. I >>>> try to add an .xml file with the same name of the text file, but still >>>> doesn't work. (I check that by first deleting all files from >>>> /context/default/populated, and populate entities from a file, and >>>> check the entities by querying the entities at >>>> http://localhost:8080/kim, but no meaningful entities found). I am >>>> wondering if I miss some steps or important configurations. Thank you >>>> very much! >>>> >>>> Best, >>>> >>>> Fangkai >>>> _______________________________________________ >>>> interested-in-KIM mailing list >>>> interested-in-KIM at ontotext.com >>>> http://ontotext.com/mailman/listinfo/interested-in-kim >>>> >>>> >>> >>> >>> >> >> >> -- >> Fangkai Yang, Ph.D student >> Taylor Hall 3.150A >> Department of Computer Sciences >> The University of Texas at Austin >> Austin, 78712-0233, Texas >> USA >> http://www.cs.utexas.edu/~fkyang >> email: fkyang at cs.utexas.edu >> >> > > > From philip.alexiev at ontotext.com Fri Jun 4 07:02:03 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Fri, 04 Jun 2010 14:02:03 +0300 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> Message-ID: <4C08DD2B.3010909@ontotext.com> Hello Fangkai, Could you send us some of your txt files that you are sure are not annotated? This could help us a lot in solving the problem. Thanks, Philip On 06/03/2010 08:00 PM, Yang Fangkai wrote: > hi, Anton, > > I tried HTML files, and the population works. But this just > doesn't work for txt file... > > I checked the populator.xml and found the following configuration: > > doc,htm,html,txt,page,xml > > I suspect the populator has already been configured to process > txt file. So where is the problem? Thank you! > > Fangkai > > 2010/6/3 Yang Fangkai: > >> Anton, >> >> On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev >> wrote: >> >>> Hello Fangkai, >>> >>> First I would like to point out that the kim-discussion: >>> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for asking >>> technical questions like this one. Next time please use the kim-discussion >>> mailing list, not this one. Thanks. >>> >>> >> Sorry for the mistake. I will use that list the next time. >> >> >>> Now back to your problem: >>> What version of KIM do you use? KIM 2.4? >>> >>> >> Yes. I am using KIM2.4 under Windows XP. >> >> >>> Are you using the KIMGate hybrid - a GATE developer with KIM's default >>> pipeline or the tool called "populater" again from the bin folder? >>> >> I started KIM by running startkim.bat, and the populator by running >> toolPopulate.cmd in tool folder. I didn't see the tool "populator" in >> the bin folder. >> >> >>> The later >>> only needs a document source folder and uses an already running KIM >>> instance. Do you see that the documents are being annotated? What results do >>> you expect, what is missing? >>> >>> >> Here is what I expect. I have a corpus containing about 2000 docs, and >> I want to query over these docs. So I plan to use toolPopulate to >> extract entities over these docs (this is what I am trying to do), and >> then query over them. I expect to see the entities populated from >> these docs, but I didn't see any meaningful entities when I query the >> entity from the KIM GUI. >> >> I don't know if the above makes sense. Thank you! >> >> Fangkai >> >> >> >>> The steps you are doing are correct in general. >>> >>> Best regards, >>> Anton Andreev >>> >>> -- >>> Anton Andreev >>> Account Manager >>> Ontotext AD >>> Tel: +359 2 875 81 17 >>> Fax:+359 2 975 32 26 >>> email: anton.andreev at ontotext.com >>> www.ontotext.com >>> >>> >>> >>> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >>> >>>> Dear List, >>>> >>>> I am trying to use Populate GUI to populate entities from my >>>> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >>>> articles from Wall Street Journal in plain text form, and refer to the >>>> folder in Populate GUI. However, it seems no entities is populated. I >>>> try to add an .xml file with the same name of the text file, but still >>>> doesn't work. (I check that by first deleting all files from >>>> /context/default/populated, and populate entities from a file, and >>>> check the entities by querying the entities at >>>> http://localhost:8080/kim, but no meaningful entities found). I am >>>> wondering if I miss some steps or important configurations. Thank you >>>> very much! >>>> >>>> Best, >>>> >>>> Fangkai >>>> _______________________________________________ >>>> interested-in-KIM mailing list >>>> interested-in-KIM at ontotext.com >>>> http://ontotext.com/mailman/listinfo/interested-in-kim >>>> >>>> >>> >>> >>> >> >> >> -- >> Fangkai Yang, Ph.D student >> Taylor Hall 3.150A >> Department of Computer Sciences >> The University of Texas at Austin >> Austin, 78712-0233, Texas >> USA >> http://www.cs.utexas.edu/~fkyang >> email: fkyang at cs.utexas.edu >> >> > > > -- Philip Alexiev Software Engineer Ontotext AD From wolfgang.yang at gmail.com Fri Jun 4 09:21:08 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Fri, 4 Jun 2010 08:21:08 -0500 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> <4C08DD2B.3010909@ontotext.com> Message-ID: Sorry I forgot to attach the file Fangkai On Fri, Jun 4, 2010 at 8:20 AM, Yang Fangkai wrote: > Hi, Philip, > > ? ? ? ? Yesterday I found a software that transformed all .txt file > to .html file and all annotation is done. However, this is not a final > solution because in the future I may have pdf or .doc file to > annotate. > > ? ? ? ? ?I am sure the attached document is not annotated. I checked > it in this way: I have a html file which contains the same content > with the .txt file, and use toolpopulate to annotate both of them, and > I use keyword "Rice University" in entity pattern search (object, > whose name is exactly equal to "Rice Univerisity"), and in the > resuult, I saw the html doc is retrieved, but .txt not. I think this > convinced me that .txt file is not annotated. > > ? ? ? ?Also, from the panel of toolpopulate, it returns the following > message after I chose .txt file to annotate: > > Checking (please wait) ... > Check: SUCCESS! > > Processing file(s) ... > > Completed: 100% ( 1 of 1 files processed ) > > Indices optimized ! > > -=[ TOTALS ]=- > Directory files: 1 > Start time: Fri Jun 04 08:13:57 CDT 2010 > End time: Fri Jun 04 08:13:57 CDT 2010 > Total time (ms): 47 > > -=[ STATISTICS ]=- > Document count: 1 > Document size (kb): 0 > Create time (ms): 0 > Parse features time (ms): 0 > Annotation time (ms): 0 > Store time (ms): 0 > Index sync time (ms): 0 > Index opt time (ms): 0 > ---------------------------------------------------------------- > End Time: Fri Jun 04 08:13:57 CDT 2010 > ---------------------------------------------------------------- > Finished. > > ? ? ? From thie message it doesn't look like the file is annotated. > > ? ? ? Thank you very much for your help! > > Fangkai > > On Fri, Jun 4, 2010 at 6:02 AM, Philip Alexiev > wrote: >> Hello Fangkai, >> >> Could you send us some of your txt files that you are sure are not >> annotated? This could help us a lot in solving the problem. >> >> Thanks, >> Philip >> >> On 06/03/2010 08:00 PM, Yang Fangkai wrote: >>> >>> hi, Anton, >>> >>> ? ? ? ? I tried HTML files, and the population works. But this just >>> doesn't work for txt file... >>> >>> ? ? ? ?I checked the populator.xml and found the following configuration: >>> >>> ? ? ? ?doc,htm,html,txt,page,xml >>> >>> ? ? ? ?I suspect the populator has already been configured to process >>> txt file. So where is the problem? Thank you! >>> >>> Fangkai >>> >>> 2010/6/3 Yang Fangkai: >>> >>>> >>>> Anton, >>>> >>>> On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev >>>> ?wrote: >>>> >>>>> >>>>> Hello Fangkai, >>>>> >>>>> First I would like to point out that the kim-discussion: >>>>> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for >>>>> asking >>>>> technical questions like this one. Next time please use the >>>>> kim-discussion >>>>> mailing list, not this one. Thanks. >>>>> >>>>> >>>> >>>> Sorry for the mistake. I will use that list the next time. >>>> >>>> >>>>> >>>>> Now back to your problem: >>>>> What version of KIM do you use? KIM 2.4? >>>>> >>>>> >>>> >>>> Yes. I am using KIM2.4 under Windows XP. >>>> >>>> >>>>> >>>>> Are you using the KIMGate hybrid - a GATE developer with KIM's default >>>>> pipeline or the tool called "populater" again from the bin folder? >>>>> >>>> >>>> I started KIM by running startkim.bat, and the populator by running >>>> toolPopulate.cmd in tool folder. I didn't see the tool "populator" in >>>> the bin folder. >>>> >>>> >>>>> >>>>> The later >>>>> only needs a document source folder and uses an already running KIM >>>>> instance. Do you see that the documents are being annotated? What >>>>> results do >>>>> you expect, what is missing? >>>>> >>>>> >>>> >>>> Here is what I expect. I have a corpus containing about 2000 docs, and >>>> I want to query over these docs. So I plan to use toolPopulate to >>>> extract entities over these docs (this is what I am trying to do), and >>>> then query over them. I expect to see the entities populated from >>>> these docs, but I didn't see any meaningful entities when I query the >>>> entity from the KIM GUI. >>>> >>>> I don't know if the above makes sense. Thank you! >>>> >>>> Fangkai >>>> >>>> >>>> >>>>> >>>>> The steps you are doing are correct in general. >>>>> >>>>> Best regards, >>>>> Anton Andreev >>>>> >>>>> -- >>>>> Anton Andreev >>>>> Account Manager >>>>> Ontotext AD >>>>> Tel: +359 2 875 81 17 >>>>> Fax:+359 2 975 32 26 >>>>> email: anton.andreev at ontotext.com >>>>> www.ontotext.com >>>>> >>>>> >>>>> >>>>> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >>>>> >>>>>> >>>>>> Dear List, >>>>>> >>>>>> ? ? ? ? ?I am trying to use Populate GUI to populate entities from my >>>>>> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >>>>>> articles from Wall Street Journal in plain text form, and refer to the >>>>>> folder in Populate GUI. However, it seems no entities is populated. I >>>>>> try to add an .xml file with the same name of the text file, but still >>>>>> doesn't work. (I check that by first deleting all files from >>>>>> /context/default/populated, and populate entities from a file, and >>>>>> check the entities by querying the entities at >>>>>> http://localhost:8080/kim, but no meaningful entities found). I am >>>>>> wondering if I miss some steps or important configurations. Thank you >>>>>> very much! >>>>>> >>>>>> Best, >>>>>> >>>>>> Fangkai >>>>>> _______________________________________________ >>>>>> interested-in-KIM mailing list >>>>>> interested-in-KIM at ontotext.com >>>>>> http://ontotext.com/mailman/listinfo/interested-in-kim >>>>>> >>>>>> >>>>> >>>>> >>>>> >>>> >>>> >>>> -- >>>> Fangkai Yang, Ph.D student >>>> Taylor Hall 3.150A >>>> Department of Computer Sciences >>>> The University of Texas at Austin >>>> Austin, 78712-0233, Texas >>>> USA >>>> http://www.cs.utexas.edu/~fkyang >>>> email: fkyang at cs.utexas.edu >>>> >>>> >>> >>> >>> >> >> >> -- >> Philip Alexiev >> Software Engineer >> Ontotext AD >> >> > > > > -- > Fangkai Yang, Ph.D student > Taylor Hall 3.150A > Department of Computer Sciences > The University of Texas at Austin > Austin, 78712-0233, Texas > USA > http://www.cs.utexas.edu/~fkyang > email: fkyang at cs.utexas.edu > -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu -------------- next part -------------- A non-text attachment was scrubbed... Name: wsj_1705 Type: application/octet-stream Size: 7137 bytes Desc: not available URL: From wolfgang.yang at gmail.com Fri Jun 4 09:20:29 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Fri, 4 Jun 2010 08:20:29 -0500 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: <4C08DD2B.3010909@ontotext.com> References: <4C07CCBE.6090906@ontotext.com> <4C08DD2B.3010909@ontotext.com> Message-ID: Hi, Philip, Yesterday I found a software that transformed all .txt file to .html file and all annotation is done. However, this is not a final solution because in the future I may have pdf or .doc file to annotate. I am sure the attached document is not annotated. I checked it in this way: I have a html file which contains the same content with the .txt file, and use toolpopulate to annotate both of them, and I use keyword "Rice University" in entity pattern search (object, whose name is exactly equal to "Rice Univerisity"), and in the resuult, I saw the html doc is retrieved, but .txt not. I think this convinced me that .txt file is not annotated. Also, from the panel of toolpopulate, it returns the following message after I chose .txt file to annotate: Checking (please wait) ... Check: SUCCESS! Processing file(s) ... Completed: 100% ( 1 of 1 files processed ) Indices optimized ! -=[ TOTALS ]=- Directory files: 1 Start time: Fri Jun 04 08:13:57 CDT 2010 End time: Fri Jun 04 08:13:57 CDT 2010 Total time (ms): 47 -=[ STATISTICS ]=- Document count: 1 Document size (kb): 0 Create time (ms): 0 Parse features time (ms): 0 Annotation time (ms): 0 Store time (ms): 0 Index sync time (ms): 0 Index opt time (ms): 0 ---------------------------------------------------------------- End Time: Fri Jun 04 08:13:57 CDT 2010 ---------------------------------------------------------------- Finished. From thie message it doesn't look like the file is annotated. Thank you very much for your help! Fangkai On Fri, Jun 4, 2010 at 6:02 AM, Philip Alexiev wrote: > Hello Fangkai, > > Could you send us some of your txt files that you are sure are not > annotated? This could help us a lot in solving the problem. > > Thanks, > Philip > > On 06/03/2010 08:00 PM, Yang Fangkai wrote: >> >> hi, Anton, >> >> ? ? ? ? I tried HTML files, and the population works. But this just >> doesn't work for txt file... >> >> ? ? ? ?I checked the populator.xml and found the following configuration: >> >> ? ? ? ?doc,htm,html,txt,page,xml >> >> ? ? ? ?I suspect the populator has already been configured to process >> txt file. So where is the problem? Thank you! >> >> Fangkai >> >> 2010/6/3 Yang Fangkai: >> >>> >>> Anton, >>> >>> On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev >>> ?wrote: >>> >>>> >>>> Hello Fangkai, >>>> >>>> First I would like to point out that the kim-discussion: >>>> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for >>>> asking >>>> technical questions like this one. Next time please use the >>>> kim-discussion >>>> mailing list, not this one. Thanks. >>>> >>>> >>> >>> Sorry for the mistake. I will use that list the next time. >>> >>> >>>> >>>> Now back to your problem: >>>> What version of KIM do you use? KIM 2.4? >>>> >>>> >>> >>> Yes. I am using KIM2.4 under Windows XP. >>> >>> >>>> >>>> Are you using the KIMGate hybrid - a GATE developer with KIM's default >>>> pipeline or the tool called "populater" again from the bin folder? >>>> >>> >>> I started KIM by running startkim.bat, and the populator by running >>> toolPopulate.cmd in tool folder. I didn't see the tool "populator" in >>> the bin folder. >>> >>> >>>> >>>> The later >>>> only needs a document source folder and uses an already running KIM >>>> instance. Do you see that the documents are being annotated? What >>>> results do >>>> you expect, what is missing? >>>> >>>> >>> >>> Here is what I expect. I have a corpus containing about 2000 docs, and >>> I want to query over these docs. So I plan to use toolPopulate to >>> extract entities over these docs (this is what I am trying to do), and >>> then query over them. I expect to see the entities populated from >>> these docs, but I didn't see any meaningful entities when I query the >>> entity from the KIM GUI. >>> >>> I don't know if the above makes sense. Thank you! >>> >>> Fangkai >>> >>> >>> >>>> >>>> The steps you are doing are correct in general. >>>> >>>> Best regards, >>>> Anton Andreev >>>> >>>> -- >>>> Anton Andreev >>>> Account Manager >>>> Ontotext AD >>>> Tel: +359 2 875 81 17 >>>> Fax:+359 2 975 32 26 >>>> email: anton.andreev at ontotext.com >>>> www.ontotext.com >>>> >>>> >>>> >>>> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >>>> >>>>> >>>>> Dear List, >>>>> >>>>> ? ? ? ? ?I am trying to use Populate GUI to populate entities from my >>>>> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >>>>> articles from Wall Street Journal in plain text form, and refer to the >>>>> folder in Populate GUI. However, it seems no entities is populated. I >>>>> try to add an .xml file with the same name of the text file, but still >>>>> doesn't work. (I check that by first deleting all files from >>>>> /context/default/populated, and populate entities from a file, and >>>>> check the entities by querying the entities at >>>>> http://localhost:8080/kim, but no meaningful entities found). I am >>>>> wondering if I miss some steps or important configurations. Thank you >>>>> very much! >>>>> >>>>> Best, >>>>> >>>>> Fangkai >>>>> _______________________________________________ >>>>> interested-in-KIM mailing list >>>>> interested-in-KIM at ontotext.com >>>>> http://ontotext.com/mailman/listinfo/interested-in-kim >>>>> >>>>> >>>> >>>> >>>> >>> >>> >>> -- >>> Fangkai Yang, Ph.D student >>> Taylor Hall 3.150A >>> Department of Computer Sciences >>> The University of Texas at Austin >>> Austin, 78712-0233, Texas >>> USA >>> http://www.cs.utexas.edu/~fkyang >>> email: fkyang at cs.utexas.edu >>> >>> >> >> >> > > > -- > Philip Alexiev > Software Engineer > Ontotext AD > > -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu From philip.alexiev at ontotext.com Fri Jun 4 09:46:12 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Fri, 04 Jun 2010 16:46:12 +0300 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> <4C08DD2B.3010909@ontotext.com> Message-ID: <4C0903A4.6050202@ontotext.com> Hi again I tried with this file and annotated it with no problems. Some things to consider: * Do your text files have .txt extension ? * Do you give the populater the file itself as a parameter, or a directory. It should be a directory. Doesn't work with files. All the best, Philip On 06/04/2010 04:21 PM, Yang Fangkai wrote: > Sorry I forgot to attach the file > > Fangkai > > On Fri, Jun 4, 2010 at 8:20 AM, Yang Fangkai wrote: > >> Hi, Philip, >> >> Yesterday I found a software that transformed all .txt file >> to .html file and all annotation is done. However, this is not a final >> solution because in the future I may have pdf or .doc file to >> annotate. >> >> I am sure the attached document is not annotated. I checked >> it in this way: I have a html file which contains the same content >> with the .txt file, and use toolpopulate to annotate both of them, and >> I use keyword "Rice University" in entity pattern search (object, >> whose name is exactly equal to "Rice Univerisity"), and in the >> resuult, I saw the html doc is retrieved, but .txt not. I think this >> convinced me that .txt file is not annotated. >> >> Also, from the panel of toolpopulate, it returns the following >> message after I chose .txt file to annotate: >> >> Checking (please wait) ... >> Check: SUCCESS! >> >> Processing file(s) ... >> >> Completed: 100% ( 1 of 1 files processed ) >> >> Indices optimized ! >> >> -=[ TOTALS ]=- >> Directory files: 1 >> Start time: Fri Jun 04 08:13:57 CDT 2010 >> End time: Fri Jun 04 08:13:57 CDT 2010 >> Total time (ms): 47 >> >> -=[ STATISTICS ]=- >> Document count: 1 >> Document size (kb): 0 >> Create time (ms): 0 >> Parse features time (ms): 0 >> Annotation time (ms): 0 >> Store time (ms): 0 >> Index sync time (ms): 0 >> Index opt time (ms): 0 >> ---------------------------------------------------------------- >> End Time: Fri Jun 04 08:13:57 CDT 2010 >> ---------------------------------------------------------------- >> Finished. >> >> From thie message it doesn't look like the file is annotated. >> >> Thank you very much for your help! >> >> Fangkai >> >> On Fri, Jun 4, 2010 at 6:02 AM, Philip Alexiev >> wrote: >> >>> Hello Fangkai, >>> >>> Could you send us some of your txt files that you are sure are not >>> annotated? This could help us a lot in solving the problem. >>> >>> Thanks, >>> Philip >>> >>> On 06/03/2010 08:00 PM, Yang Fangkai wrote: >>> >>>> hi, Anton, >>>> >>>> I tried HTML files, and the population works. But this just >>>> doesn't work for txt file... >>>> >>>> I checked the populator.xml and found the following configuration: >>>> >>>> doc,htm,html,txt,page,xml >>>> >>>> I suspect the populator has already been configured to process >>>> txt file. So where is the problem? Thank you! >>>> >>>> Fangkai >>>> >>>> 2010/6/3 Yang Fangkai: >>>> >>>> >>>>> Anton, >>>>> >>>>> On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev >>>>> wrote: >>>>> >>>>> >>>>>> Hello Fangkai, >>>>>> >>>>>> First I would like to point out that the kim-discussion: >>>>>> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for >>>>>> asking >>>>>> technical questions like this one. Next time please use the >>>>>> kim-discussion >>>>>> mailing list, not this one. Thanks. >>>>>> >>>>>> >>>>>> >>>>> Sorry for the mistake. I will use that list the next time. >>>>> >>>>> >>>>> >>>>>> Now back to your problem: >>>>>> What version of KIM do you use? KIM 2.4? >>>>>> >>>>>> >>>>>> >>>>> Yes. I am using KIM2.4 under Windows XP. >>>>> >>>>> >>>>> >>>>>> Are you using the KIMGate hybrid - a GATE developer with KIM's default >>>>>> pipeline or the tool called "populater" again from the bin folder? >>>>>> >>>>>> >>>>> I started KIM by running startkim.bat, and the populator by running >>>>> toolPopulate.cmd in tool folder. I didn't see the tool "populator" in >>>>> the bin folder. >>>>> >>>>> >>>>> >>>>>> The later >>>>>> only needs a document source folder and uses an already running KIM >>>>>> instance. Do you see that the documents are being annotated? What >>>>>> results do >>>>>> you expect, what is missing? >>>>>> >>>>>> >>>>>> >>>>> Here is what I expect. I have a corpus containing about 2000 docs, and >>>>> I want to query over these docs. So I plan to use toolPopulate to >>>>> extract entities over these docs (this is what I am trying to do), and >>>>> then query over them. I expect to see the entities populated from >>>>> these docs, but I didn't see any meaningful entities when I query the >>>>> entity from the KIM GUI. >>>>> >>>>> I don't know if the above makes sense. Thank you! >>>>> >>>>> Fangkai >>>>> >>>>> >>>>> >>>>> >>>>>> The steps you are doing are correct in general. >>>>>> >>>>>> Best regards, >>>>>> Anton Andreev >>>>>> >>>>>> -- >>>>>> Anton Andreev >>>>>> Account Manager >>>>>> Ontotext AD >>>>>> Tel: +359 2 875 81 17 >>>>>> Fax:+359 2 975 32 26 >>>>>> email: anton.andreev at ontotext.com >>>>>> www.ontotext.com >>>>>> >>>>>> >>>>>> >>>>>> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >>>>>> >>>>>> >>>>>>> Dear List, >>>>>>> >>>>>>> I am trying to use Populate GUI to populate entities from my >>>>>>> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >>>>>>> articles from Wall Street Journal in plain text form, and refer to the >>>>>>> folder in Populate GUI. However, it seems no entities is populated. I >>>>>>> try to add an .xml file with the same name of the text file, but still >>>>>>> doesn't work. (I check that by first deleting all files from >>>>>>> /context/default/populated, and populate entities from a file, and >>>>>>> check the entities by querying the entities at >>>>>>> http://localhost:8080/kim, but no meaningful entities found). I am >>>>>>> wondering if I miss some steps or important configurations. Thank you >>>>>>> very much! >>>>>>> >>>>>>> Best, >>>>>>> >>>>>>> Fangkai >>>>>>> _______________________________________________ >>>>>>> interested-in-KIM mailing list >>>>>>> interested-in-KIM at ontotext.com >>>>>>> http://ontotext.com/mailman/listinfo/interested-in-kim >>>>>>> >>>>>>> >>>>>>> >>>>>> >>>>>> >>>>>> >>>>> >>>>> -- >>>>> Fangkai Yang, Ph.D student >>>>> Taylor Hall 3.150A >>>>> Department of Computer Sciences >>>>> The University of Texas at Austin >>>>> Austin, 78712-0233, Texas >>>>> USA >>>>> http://www.cs.utexas.edu/~fkyang >>>>> email: fkyang at cs.utexas.edu >>>>> >>>>> >>>>> >>>> >>>> >>>> >>> >>> -- >>> Philip Alexiev >>> Software Engineer >>> Ontotext AD >>> >>> >>> >> >> >> -- >> Fangkai Yang, Ph.D student >> Taylor Hall 3.150A >> Department of Computer Sciences >> The University of Texas at Austin >> Austin, 78712-0233, Texas >> USA >> http://www.cs.utexas.edu/~fkyang >> email: fkyang at cs.utexas.edu >> >> > > > > > > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion > -- Philip Alexiev Software Engineer Ontotext AD -------------- next part -------------- An HTML attachment was scrubbed... URL: From wolfgang.yang at gmail.com Fri Jun 4 09:54:04 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Fri, 4 Jun 2010 08:54:04 -0500 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: <4C0903A4.6050202@ontotext.com> References: <4C07CCBE.6090906@ontotext.com> <4C08DD2B.3010909@ontotext.com> <4C0903A4.6050202@ontotext.com> Message-ID: Ah.....yeah, it works now. I think I made mistakes in both of the points you mentioned. Thank you so much for your help! Fangkai On Fri, Jun 4, 2010 at 8:46 AM, Philip Alexiev wrote: > Hi again > > > I tried with this file and annotated it with no problems. > > Some things to consider: > > * Do your text files have? .txt? extension ? > * Do you give the populater the file itself as a parameter, or a directory. > It should be a directory. Doesn't work with files. > > All the best, > Philip > > > On 06/04/2010 04:21 PM, Yang Fangkai wrote: > > Sorry I forgot to attach the file > Fangkai > On Fri, Jun 4, 2010 at 8:20 AM, Yang Fangkai > wrote: > > > Hi, Philip, > ? ? ? ? Yesterday I found a software that transformed all .txt file > to .html file and all annotation is done. However, this is not a final > solution because in the future I may have pdf or .doc file to > annotate. > ? ? ? ? ?I am sure the attached document is not annotated. I checked > it in this way: I have a html file which contains the same content > with the .txt file, and use toolpopulate to annotate both of them, and > I use keyword "Rice University" in entity pattern search (object, > whose name is exactly equal to "Rice Univerisity"), and in the > resuult, I saw the html doc is retrieved, but .txt not. I think this > convinced me that .txt file is not annotated. > ? ? ? ?Also, from the panel of toolpopulate, it returns the following > message after I chose .txt file to annotate: > Checking (please wait) ... > Check: SUCCESS! > Processing file(s) ... > Completed: 100% ( 1 of 1 files processed ) > Indices optimized ! > -=[ TOTALS ]=- > Directory files: 1 > Start time: Fri Jun 04 08:13:57 CDT 2010 > End time: Fri Jun 04 08:13:57 CDT 2010 > Total time (ms): 47 > -=[ STATISTICS ]=- > Document count: 1 > Document size (kb): 0 > Create time (ms): 0 > Parse features time (ms): 0 > Annotation time (ms): 0 > Store time (ms): 0 > Index sync time (ms): 0 > Index opt time (ms): 0 > ---------------------------------------------------------------- > End Time: Fri Jun 04 08:13:57 CDT 2010 > ---------------------------------------------------------------- > Finished. > ? ? ? From thie message it doesn't look like the file is annotated. > ? ? ? Thank you very much for your help! > Fangkai > On Fri, Jun 4, 2010 at 6:02 AM, Philip Alexiev > wrote: > > > Hello Fangkai, > Could you send us some of your txt files that you are sure are not > annotated? This could help us a lot in solving the problem. > Thanks, > Philip > On 06/03/2010 08:00 PM, Yang Fangkai wrote: > > > hi, Anton, > ? ? ? ? I tried HTML files, and the population works. But this just > doesn't work for txt file... > ? ? ? ?I checked the populator.xml and found the following configuration: > ? ? ? ?doc,htm,html,txt,page,xml > ? ? ? ?I suspect the populator has already been configured to process > txt file. So where is the problem? Thank you! > Fangkai > 2010/6/3 Yang Fangkai: > > > Anton, > On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev > ?wrote: > > > Hello Fangkai, > First I would like to point out that the kim-discussion: > http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for > asking > technical questions like this one. Next time please use the > kim-discussion > mailing list, not this one. Thanks. > > > Sorry for the mistake. I will use that list the next time. > > > Now back to your problem: > What version of KIM do you use? KIM 2.4? > > > Yes. I am using KIM2.4 under Windows XP. > > > Are you using the KIMGate hybrid - a GATE developer with KIM's default > pipeline or the tool called "populater" again from the bin folder? > > > I started KIM by running startkim.bat, and the populator by running > toolPopulate.cmd in tool folder. I didn't see the tool "populator" in > the bin folder. > > > The later > only needs a document source folder and uses an already running KIM > instance. Do you see that the documents are being annotated? What > results do > you expect, what is missing? > > > Here is what I expect. I have a corpus containing about 2000 docs, and > I want to query over these docs. So I plan to use toolPopulate to > extract entities over these docs (this is what I am trying to do), and > then query over them. I expect to see the entities populated from > these docs, but I didn't see any meaningful entities when I query the > entity from the KIM GUI. > I don't know if the above makes sense. Thank you! > Fangkai > > > The steps you are doing are correct in general. > Best regards, > Anton Andreev > -- > Anton Andreev > Account Manager > Ontotext AD > Tel: +359 2 875 81 17 > Fax:+359 2 975 32 26 > email: anton.andreev at ontotext.com > www.ontotext.com > On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: > > > Dear List, > ? ? ? ? ?I am trying to use Populate GUI to populate entities from my > own corpus. I have downloaded the raw file of PennTree bank, i.e., the > articles from Wall Street Journal in plain text form, and refer to the > folder in Populate GUI. However, it seems no entities is populated. I > try to add an .xml file with the same name of the text file, but still > doesn't work. (I check that by first deleting all files from > /context/default/populated, and populate entities from a file, and > check the entities by querying the entities at > http://localhost:8080/kim, but no meaningful entities found). I am > wondering if I miss some steps or important configurations. Thank you > very much! > Best, > Fangkai > _______________________________________________ > interested-in-KIM mailing list > interested-in-KIM at ontotext.com > http://ontotext.com/mailman/listinfo/interested-in-kim > > > > > -- > Fangkai Yang, Ph.D student > Taylor Hall 3.150A > Department of Computer Sciences > The University of Texas at Austin > Austin, 78712-0233, Texas > USA > http://www.cs.utexas.edu/~fkyang > email: fkyang at cs.utexas.edu > > > > > -- > Philip Alexiev > Software Engineer > Ontotext AD > > > -- > Fangkai Yang, Ph.D student > Taylor Hall 3.150A > Department of Computer Sciences > The University of Texas at Austin > Austin, 78712-0233, Texas > USA > http://www.cs.utexas.edu/~fkyang > email: fkyang at cs.utexas.edu > > > > > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion > > > -- > Philip Alexiev > Software Engineer > Ontotext AD -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu From philip.alexiev at ontotext.com Fri Jun 4 10:33:26 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Fri, 04 Jun 2010 17:33:26 +0300 Subject: [Kim-discussion] [Interested-in-kim] problem of populating instances from my own corpus In-Reply-To: References: <4C07CCBE.6090906@ontotext.com> <4C08DD2B.3010909@ontotext.com> <4C0903A4.6050202@ontotext.com> Message-ID: <4C090EB6.10103@ontotext.com> Happy that we could help :) All the best, Philip On 06/04/2010 04:54 PM, Yang Fangkai wrote: > Ah.....yeah, it works now. > > I think I made mistakes in both of the points you mentioned. > > Thank you so much for your help! > > Fangkai > > On Fri, Jun 4, 2010 at 8:46 AM, Philip Alexiev > wrote: > >> Hi again >> >> >> I tried with this file and annotated it with no problems. >> >> Some things to consider: >> >> * Do your text files have .txt extension ? >> * Do you give the populater the file itself as a parameter, or a directory. >> It should be a directory. Doesn't work with files. >> >> All the best, >> Philip >> >> >> On 06/04/2010 04:21 PM, Yang Fangkai wrote: >> >> Sorry I forgot to attach the file >> Fangkai >> On Fri, Jun 4, 2010 at 8:20 AM, Yang Fangkai >> wrote: >> >> >> Hi, Philip, >> Yesterday I found a software that transformed all .txt file >> to .html file and all annotation is done. However, this is not a final >> solution because in the future I may have pdf or .doc file to >> annotate. >> I am sure the attached document is not annotated. I checked >> it in this way: I have a html file which contains the same content >> with the .txt file, and use toolpopulate to annotate both of them, and >> I use keyword "Rice University" in entity pattern search (object, >> whose name is exactly equal to "Rice Univerisity"), and in the >> resuult, I saw the html doc is retrieved, but .txt not. I think this >> convinced me that .txt file is not annotated. >> Also, from the panel of toolpopulate, it returns the following >> message after I chose .txt file to annotate: >> Checking (please wait) ... >> Check: SUCCESS! >> Processing file(s) ... >> Completed: 100% ( 1 of 1 files processed ) >> Indices optimized ! >> -=[ TOTALS ]=- >> Directory files: 1 >> Start time: Fri Jun 04 08:13:57 CDT 2010 >> End time: Fri Jun 04 08:13:57 CDT 2010 >> Total time (ms): 47 >> -=[ STATISTICS ]=- >> Document count: 1 >> Document size (kb): 0 >> Create time (ms): 0 >> Parse features time (ms): 0 >> Annotation time (ms): 0 >> Store time (ms): 0 >> Index sync time (ms): 0 >> Index opt time (ms): 0 >> ---------------------------------------------------------------- >> End Time: Fri Jun 04 08:13:57 CDT 2010 >> ---------------------------------------------------------------- >> Finished. >> From thie message it doesn't look like the file is annotated. >> Thank you very much for your help! >> Fangkai >> On Fri, Jun 4, 2010 at 6:02 AM, Philip Alexiev >> wrote: >> >> >> Hello Fangkai, >> Could you send us some of your txt files that you are sure are not >> annotated? This could help us a lot in solving the problem. >> Thanks, >> Philip >> On 06/03/2010 08:00 PM, Yang Fangkai wrote: >> >> >> hi, Anton, >> I tried HTML files, and the population works. But this just >> doesn't work for txt file... >> I checked the populator.xml and found the following configuration: >> doc,htm,html,txt,page,xml >> I suspect the populator has already been configured to process >> txt file. So where is the problem? Thank you! >> Fangkai >> 2010/6/3 Yang Fangkai: >> >> >> Anton, >> On Thu, Jun 3, 2010 at 10:39 AM, Anton Andreev >> wrote: >> >> >> Hello Fangkai, >> First I would like to point out that the kim-discussion: >> http://ontotext.com/mailman/listinfo/kim-discussion is dedicated for >> asking >> technical questions like this one. Next time please use the >> kim-discussion >> mailing list, not this one. Thanks. >> >> >> Sorry for the mistake. I will use that list the next time. >> >> >> Now back to your problem: >> What version of KIM do you use? KIM 2.4? >> >> >> Yes. I am using KIM2.4 under Windows XP. >> >> >> Are you using the KIMGate hybrid - a GATE developer with KIM's default >> pipeline or the tool called "populater" again from the bin folder? >> >> >> I started KIM by running startkim.bat, and the populator by running >> toolPopulate.cmd in tool folder. I didn't see the tool "populator" in >> the bin folder. >> >> >> The later >> only needs a document source folder and uses an already running KIM >> instance. Do you see that the documents are being annotated? What >> results do >> you expect, what is missing? >> >> >> Here is what I expect. I have a corpus containing about 2000 docs, and >> I want to query over these docs. So I plan to use toolPopulate to >> extract entities over these docs (this is what I am trying to do), and >> then query over them. I expect to see the entities populated from >> these docs, but I didn't see any meaningful entities when I query the >> entity from the KIM GUI. >> I don't know if the above makes sense. Thank you! >> Fangkai >> >> >> The steps you are doing are correct in general. >> Best regards, >> Anton Andreev >> -- >> Anton Andreev >> Account Manager >> Ontotext AD >> Tel: +359 2 875 81 17 >> Fax:+359 2 975 32 26 >> email: anton.andreev at ontotext.com >> www.ontotext.com >> On 3.6.2010 ?. 18:17 ?., KIM Platform info newsletter wrote: >> >> >> Dear List, >> I am trying to use Populate GUI to populate entities from my >> own corpus. I have downloaded the raw file of PennTree bank, i.e., the >> articles from Wall Street Journal in plain text form, and refer to the >> folder in Populate GUI. However, it seems no entities is populated. I >> try to add an .xml file with the same name of the text file, but still >> doesn't work. (I check that by first deleting all files from >> /context/default/populated, and populate entities from a file, and >> check the entities by querying the entities at >> http://localhost:8080/kim, but no meaningful entities found). I am >> wondering if I miss some steps or important configurations. Thank you >> very much! >> Best, >> Fangkai >> _______________________________________________ >> interested-in-KIM mailing list >> interested-in-KIM at ontotext.com >> http://ontotext.com/mailman/listinfo/interested-in-kim >> >> >> >> >> -- >> Fangkai Yang, Ph.D student >> Taylor Hall 3.150A >> Department of Computer Sciences >> The University of Texas at Austin >> Austin, 78712-0233, Texas >> USA >> http://www.cs.utexas.edu/~fkyang >> email: fkyang at cs.utexas.edu >> >> >> >> >> -- >> Philip Alexiev >> Software Engineer >> Ontotext AD >> >> >> -- >> Fangkai Yang, Ph.D student >> Taylor Hall 3.150A >> Department of Computer Sciences >> The University of Texas at Austin >> Austin, 78712-0233, Texas >> USA >> http://www.cs.utexas.edu/~fkyang >> email: fkyang at cs.utexas.edu >> >> >> >> >> _______________________________________________ >> Kim-discussion mailing list >> Kim-discussion at ontotext.com >> http://ontotext.com/mailman/listinfo/kim-discussion >> >> >> -- >> Philip Alexiev >> Software Engineer >> Ontotext AD >> > > > -- Philip Alexiev Software Engineer Ontotext AD From wolfgang.yang at gmail.com Fri Jun 4 16:05:55 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Fri, 4 Jun 2010 15:05:55 -0500 Subject: [Kim-discussion] questions about trusted entities and extending ontologies Message-ID: Dear List, I have three questions: 1. After entities are populated from corpus by toolPopulate GUI, will these entities be treated as "trusted entities" for further document retrieval? I didn't extend either ontology or KB, but used all default setting of KIM2.4. I know KIM treats trusted (pre-populated) entities and recognized entities differently, but what kind of entities does those populated from corpus belong to? In what way does KIM treat them differently? 2. As as mentioned here http://ontotext.com/kim/doc/sys-doc/DesigningOntology.html, extending KIM ontology with new ones needs to define the new ontology based on PROTON. However, my problem now is I have a huge and complicated domain ontology written in OWL. Does this mean I should re-analyze the whole domain ontology and manually align each class/properties into PROTON? Is there any easy way to do that? 3. I used Protege to open protons.owl, but it returned the following error message. I am wondering why... (maybe this is no longer the KIM question, but I will be very thankful if you could offer me a few hints) org.semanticweb.owlapi.rdf.syntax.RDFParserException: [line=92:column=48] IRI 'http://www.w3.org/2000/01/rdf-schema##Literal' cannot be resolved against curent base IRI http://proton.semanticweb.org/2006/05/protons# Thank you! Fangkai From wolfgang.yang at gmail.com Fri Jun 4 16:20:56 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Fri, 4 Jun 2010 15:20:56 -0500 Subject: [Kim-discussion] questions about trusted entities and extending ontologies In-Reply-To: References: Message-ID: Dear List, On Fri, Jun 4, 2010 at 3:05 PM, Yang Fangkai wrote: > Dear List, > > ? ? ? ? I have three questions: > > ? ? ? 1. ?After entities are populated from corpus by toolPopulate > GUI, will these entities be treated as "trusted entities" for further > document retrieval? ?I didn't extend either ontology or KB, but used > all default setting of KIM2.4. I know KIM treats trusted > (pre-populated) entities and recognized entities differently, but what > kind of entities does those populated from corpus belong to? In what > way does KIM treat them differently? > This is a naive question. I have found the answer to these. Please disregard this question. Fangkai From philip.alexiev at ontotext.com Mon Jun 7 04:41:59 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Mon, 07 Jun 2010 11:41:59 +0300 Subject: [Kim-discussion] questions about trusted entities and extending ontologies In-Reply-To: References: Message-ID: <4C0CB0D7.2000801@ontotext.com> .Hi Yang, On 06/04/2010 11:05 PM, Yang Fangkai wrote: > 2. As mentioned here > http://ontotext.com/kim/doc/sys-doc/DesigningOntology.html, extending > KIM ontology with new ones needs to define the new ontology based on > PROTON. However, my problem now is I have a huge and complicated > domain ontology written in OWL. Does this mean I should re-analyze the > whole domain ontology and manually align each class/properties into > PROTON? Is there any easy way to do that? > I am not aware of any automated and/or easier way to do this. Follow the instructions in documentation ( http://ontotext.com/kim/doc/sys-doc/DesigningOntology.html ) and especially in 3. Integrating Ontology Extensions. The important points are that if you want your objects to be recognized in the texts, you should follow some rules: 1. All the classes, the objects belong to, should inherit directly or indirectly http://proton.semanticweb.org/2006/05/protons#Entity. 2. Any added classes or properties (relations) must be declared visible. Those rules are what the LKB Gazetteer resource uses to fill its lists from the ontology. So following them will result in the gazetteer using the objects for recognition. 3. It is also good to be more specific with the inheritance and whenever possible inherit some of the more specific classes in PROTON Top module. This also servers as a base for the rules in the Jape Grammars. > 3. I used Protege to open protons.owl, but it returned the > following error message. I am wondering why... (maybe this is no > longer the KIM question, but I will be very thankful if you could > offer me a few hints) > > org.semanticweb.owlapi.rdf.syntax.RDFParserException: > [line=92:column=48] IRI > 'http://www.w3.org/2000/01/rdf-schema##Literal' cannot be resolved > against curent base IRIhttp://proton.semanticweb.org/2006/05/protons# > Never came across this error. But you may try some other ontology editors/viewers like Swoop or TopBraid Composer . Greetings, Philip -- Philip Alexiev Software Engineer Ontotext AD -------------- next part -------------- An HTML attachment was scrubbed... URL: From alistair.duke at bt.com Mon Jun 7 04:58:30 2010 From: alistair.duke at bt.com (alistair.duke at bt.com) Date: Mon, 7 Jun 2010 09:58:30 +0100 Subject: [Kim-discussion] questions about trusted entities and extending ontologies In-Reply-To: <4C0CB0D7.2000801@ontotext.com> References: <4C0CB0D7.2000801@ontotext.com> Message-ID: Hi Yang, Are you behind a proxy? I?ve come across this error when prot?g? can?t see beyond the proxy. See: https://mailman.stanford.edu/pipermail/protege-discussion/2007-February/000735.html for information on how to resolve this. Alistair Alistair Duke Future Business Applications & Services BT Innovate & Design ___________________________ Office: +44(0) 1473 608192 Mobile: +44(0) 7730 426257 Fax : +44(0) 1473 606755 Email: alistair.duke at bt.com This email contains BT information, which may be privileged or confidential. It's meant only for the individual(s) or entity named above. If you're not the intended recipient, note that disclosing, copying, distributing or using this information is prohibited. If you've received this email in error, please let me know immediately on the email address above. Thank you. We monitor our email system, and may record your emails. British Telecommunications plc Registered office: 81 Newgate Street London EC1A 7AJ Registered in England no: 1800000 From: kim-discussion-bounces at ontotext.com [mailto:kim-discussion-bounces at ontotext.com] On Behalf Of Philip Alexiev Sent: 07 June 2010 09:42 To: Yang Fangkai Cc: kim-discussion at ontotext.com Subject: Re: [Kim-discussion] questions about trusted entities and extending ontologies .Hi Yang, On 06/04/2010 11:05 PM, Yang Fangkai wrote: 2. As mentioned here http://ontotext.com/kim/doc/sys-doc/DesigningOntology.html, extending KIM ontology with new ones needs to define the new ontology based on PROTON. However, my problem now is I have a huge and complicated domain ontology written in OWL. Does this mean I should re-analyze the whole domain ontology and manually align each class/properties into PROTON? Is there any easy way to do that? I am not aware of any automated and/or easier way to do this. Follow the instructions in documentation ( http://ontotext.com/kim/doc/sys-doc/DesigningOntology.html ) and especially in 3. Integrating Ontology Extensions. The important points are that if you want your objects to be recognized in the texts, you should follow some rules: 1. All the classes, the objects belong to, should inherit directly or indirectly http://proton.semanticweb.org/2006/05/protons#Entity. 2. Any added classes or properties (relations) must be declared visible. Those rules are what the LKB Gazetteer resource uses to fill its lists from the ontology. So following them will result in the gazetteer using the objects for recognition. 3. It is also good to be more specific with the inheritance and whenever possible inherit some of the more specific classes in PROTON Top module. This also servers as a base for the rules in the Jape Grammars. 3. I used Protege to open protons.owl, but it returned the following error message. I am wondering why... (maybe this is no longer the KIM question, but I will be very thankful if you could offer me a few hints) org.semanticweb.owlapi.rdf.syntax.RDFParserException: [line=92:column=48] IRI 'http://www.w3.org/2000/01/rdf-schema##Literal' cannot be resolved against curent base IRI http://proton.semanticweb.org/2006/05/protons# Never came across this error. But you may try some other ontology editors/viewers like Swoop or TopBraid Composer . Greetings, Philip -- Philip Alexiev Software Engineer Ontotext AD -------------- next part -------------- An HTML attachment was scrubbed... URL: From philip.alexiev at ontotext.com Mon Jun 7 05:03:15 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Mon, 07 Jun 2010 12:03:15 +0300 Subject: [Kim-discussion] questions about trusted entities and extending ontologies In-Reply-To: References: Message-ID: <4C0CB5D3.2090208@ontotext.com> Hi Again On 06/04/2010 11:05 PM, Yang Fangkai wrote: > 3. I used Protege to open protons.owl, but it returned the > following error message. I am wondering why... (maybe this is no > longer the KIM question, but I will be very thankful if you could > offer me a few hints) > > org.semanticweb.owlapi.rdf.syntax.RDFParserException: > [line=92:column=48] IRI > 'http://www.w3.org/2000/01/rdf-schema##Literal' cannot be resolved > against curent base IRIhttp://proton.semanticweb.org/2006/05/protons# > > The problem is in the two '#' symbols in the URI . This usually happens when the namespace is defined with an '#' and the class definition contains '#' as well. For example: <----- the use of # here is a flaw The definitions are correct in protons. I guess you have modified something or added something besides protons. Hope this helps Philip -- Philip Alexiev Software Engineer Ontotext AD From david.c.harrill at lmco.com Wed Jun 9 11:49:40 2010 From: david.c.harrill at lmco.com (Harrill, David C) Date: Wed, 09 Jun 2010 11:49:40 -0400 Subject: [Kim-discussion] GATE_LOCATION_DISAMBIGUATION Message-ID: <252DC4CD40D8754E9F4E258E84BCD2975A2AA6B7@HVXMSP2.us.lmco.com> To whom it may concern, I have been researching methods at which to perform disambiguation for Location Entities within GATE. As an example, I am searching for a method to differentiate Paris, France from Paris, Virginia. I spoke with one of the GATE developers who stated that Ontotext has successfully performed this task within KIM. Is there any information that you can provide me in order to perform disambiguation within GATE. Additionally, is KIM performing this function within an ontology? I would appreciate any information you could provide in regard to this matter. Thanks, Dave -------------- next part -------------- An HTML attachment was scrubbed... URL: From Anton.Andreev at ontotext.com Thu Jun 10 09:52:48 2010 From: Anton.Andreev at ontotext.com (Anton Andreev) Date: Thu, 10 Jun 2010 16:52:48 +0300 Subject: [Kim-discussion] GATE_LOCATION_DISAMBIGUATION In-Reply-To: <252DC4CD40D8754E9F4E258E84BCD2975A2AA6B7@HVXMSP2.us.lmco.com> References: <252DC4CD40D8754E9F4E258E84BCD2975A2AA6B7@HVXMSP2.us.lmco.com> Message-ID: <4C10EE30.90601@ontotext.com> Hello Mr. Harrill, Our colleagues in Sheffield are correct that we have experience in solving such problems. However, there is no standard component that can solve such disambiguation, because the task may have various definitions. A task where we have "Paris, France" in the text and we want to match it with Paris in an ontology like DBPedia, is doable in GATE using the IdRF framework from Ontotext. By "match" I mean to assign to this Paris the correct DBPedia URI. If you are trying to solve that simple, but not trivial task, Ontotext can provide documentation and, if needed, services for using IdRF in GATE. The second, more general, case is when "Paris" is mentioned in the same document with names like "France" or "Nicolas Sarkozy" which point to a specific unambiguous location, but they are not used next to "Paris". We have expertise in this area, but we have only developed domain-specific solutions, like disambiguating protein abbreviations in the biomedical domain. See [1] for an example for a possible approach. Note that the spreading activation mentioned by that research paper is supported by our BigOWLIM database. We will be happy to provide services for developing a GATE component based on that or similar approach for named entity disambiguation. Best regards, Anton Andreev [1] http://www.springerlink.com/content/f8856804480h2624/ -- Anton Andreev Account Manager Ontotext AD Tel: +359 2 875 81 17 Fax:+359 2 975 32 26 email: anton.andreev at ontotext.com www.ontotext.com On 9.6.2010 ?. 18:49 ?., Harrill, David C wrote: > > To whom it may concern, > > I have been researching methods at which to perform disambiguation for > Location Entities within GATE. As an example, I am searching for a > method to differentiate Paris, France from Paris, Virginia. I spoke > with one of the GATE developers who stated that Ontotext has > successfully performed this task within KIM. Is there any information > that you can provide me in order to perform disambiguation within > GATE. Additionally, is KIM performing this function within an > ontology? I would appreciate any information you could provide in > regard to this matter. > > Thanks, > > Dave > > > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion > -------------- next part -------------- An HTML attachment was scrubbed... URL: From borislav.popov at ontotext.com Tue Jun 15 12:42:33 2010 From: borislav.popov at ontotext.com (borislav popov) Date: Tue, 15 Jun 2010 19:42:33 +0300 Subject: [Kim-discussion] Update In-Reply-To: <1783F19CFAA5694C83FB8D718346C5BC0756115E@S-DC-EXM12.net1.cec.eu.int> References: <1783F19CFAA5694C83FB8D718346C5BC0756115D@S-DC-EXM12.net1.cec.eu.int> <4C178C31.4020403@ontotext.com> <1783F19CFAA5694C83FB8D718346C5BC0756115E@S-DC-EXM12.net1.cec.eu.int> Message-ID: <54A59B0C-A0AE-434D-99C0-DE6E5BBC2BCE@ontotext.com> i'm forwarding this to philip. my tech knowledge about kim configs is a bit superficial nowadays. he will help you and will also observe the log of the communication so far. all the best b On Jun 15, 2010, at 6:15 PM, wrote: > Borislav, > > Can you help me with this? > > I have my path pointing at the relevant Java SDK (jdk 1.6.0_14). > I have checked that config_machine has JAVA_HOME (jdk 1.6.0_14) and > TOMCAT_HOME right. > I have changed startTomcat.bat to call for tomcat6.exe in stead of > startup.bat (which doesn't exist) > > Then I start KIM by invoking startKIM.bat and I get the output as in > attached file: StartKim.rtf > > After that I start the populator by invoking toolpopulator.bat and > get the output in file "Populator.rtf" > > Finally I start Tomcat by invoking startTomcat.bat and get the > output in file "Tomcat.rtf" > > When I open my IE and try localhost:8080, I get the usual KIM web > user interface but it doesn't provide me any results. I get the > message: "ready, with errors on the page" > > Have I missed something or what. I tried to follow the KIM guides as > best I could... > > Best regards, > > Juha > > > From: Anton Andreev [mailto:Anton.Andreev at ontotext.com] > Sent: Tuesday, June 15, 2010 3:21 PM > To: JUNTTILA Juha (SANCO) > Cc: borislav popov > Subject: Re: Update > > Hello Juha, > > Nice to hear that you had some time to work with KIM. > > Borislav (see cc) will assist you further while you explore KIM. > Please address your next emails to him. > > Best regards, > Anton Andreev > -- > Anton Andreev > Account Manager > Ontotext AD > Tel: +359 2 875 81 17 > Fax:+359 2 975 32 26 > email: anton.andreev at ontotext.com > www.ontotext.com > > > On 15.6.2010 ?. 17:14 ?., Juha.Junttila at ec.europa.eu wrote: >> >> Anton, >> >> I have now been working with my KIM-installation but it seems to be >> an up-hill struggle. I manage to start KIM and Tomcat and >> apparently populate the repository but I don't get anything out >> from the web-ui. I have tried to populate the corpus with the >> example texts provided with KIM. I thought it would be a good idea >> to start with the existing application first and once it is up and >> running I can start modifying it. But it just doesn't work for me. >> >> Maybe I need to start from scratch again and re-install KIM, Java >> and Tomcat... >> >> From the documentation everything seemed fairly simple but now my >> head is full of question-marks. >> >> Do you have any great ideas? >> >> Cheers, >> >> Juha >> >> P.S. my last attempt resulted in populator tool disappear. It >> simply doesn't start and I am trying now by re-starting my lap-top. >> > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From philip.alexiev at ontotext.com Wed Jun 16 03:37:40 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Wed, 16 Jun 2010 10:37:40 +0300 Subject: [Kim-discussion] Update In-Reply-To: <54A59B0C-A0AE-434D-99C0-DE6E5BBC2BCE@ontotext.com> References: <1783F19CFAA5694C83FB8D718346C5BC0756115D@S-DC-EXM12.net1.cec.eu.int> <4C178C31.4020403@ontotext.com> <1783F19CFAA5694C83FB8D718346C5BC0756115E@S-DC-EXM12.net1.cec.eu.int> <54A59B0C-A0AE-434D-99C0-DE6E5BBC2BCE@ontotext.com> Message-ID: <4C187F44.1030105@ontotext.com> Hello Juha, I have looked at the logs, and everything seems to be OK. It looks to me like a browser/tomcat issue. Please try the actions below: * try to open not localhost:8080 but http://localhost:8080/KIM/ (you should have deployed KIM.war to $TOMCAT/webapps/ by now) * if this does not help - remove tomcat's cache : $TOMCAT/work/* and $TOMCAT/temp/* and try again * if this does not help - restart tomcat and try again Hope this helps, Philip On 06/15/2010 07:42 PM, borislav popov wrote: > i'm forwarding this to philip. my tech knowledge about kim configs is > a bit superficial nowadays. > he will help you and will also observe the log of the communication so > far. > all the best > b > > On Jun 15, 2010, at 6:15 PM, > wrote: > >> Borislav, >> Can you help me with this? >> I have my path pointing at the relevant Java SDK (jdk 1.6.0_14). >> I have checked that config_machine has JAVA_HOME (jdk 1.6.0_14) and >> TOMCAT_HOME right. >> I have changed startTomcat.bat to call for tomcat6.exe in stead of >> startup.bat (which doesn't exist) >> Then I start KIM by invoking startKIM.bat and I get the output as in >> attached file: StartKim.rtf >> After that I start the populator by invoking toolpopulator.bat and >> get the output in file "Populator.rtf" >> Finally I start Tomcat by invoking startTomcat.bat and get the output >> in file "Tomcat.rtf" >> When I open my IE and try localhost:8080, I get the usual KIM web >> user interface but it doesn't provide me any results. I get the >> message: "ready, with errors on the page" >> Have I missed something or what. I tried to follow the KIM guides as >> best I could... >> Best regards, >> Juha >> >> ------------------------------------------------------------------------ >> *From:* Anton Andreev [mailto:Anton.Andreev at ontotext.com] >> *Sent:* Tuesday, June 15, 2010 3:21 PM >> *To:* JUNTTILA Juha (SANCO) >> *Cc:* borislav popov >> *Subject:* Re: Update >> >> Hello Juha, >> >> Nice to hear that you had some time to work with KIM. >> >> Borislav (see cc) will assist you further while you explore KIM. >> Please address your next emails to him. >> >> Best regards, >> Anton Andreev >> >> -- >> Anton Andreev >> Account Manager >> Ontotext AD >> Tel: +359 2 875 81 17 >> Fax:+359 2 975 32 26 >> email:anton.andreev at ontotext.com >> www.ontotext.com >> >> >> >> On 15.6.2010 ?. 17:14 ?., Juha.Junttila at ec.europa.eu wrote: >>> >>> Anton, >>> >>> I have now been working with my KIM-installation but it seems to >>> be an up-hill struggle. I manage to start KIM and Tomcat and >>> apparently populate the repository but I don't get anything out >>> from the web-ui. I have tried to populate the corpus with the >>> example texts provided with KIM. I thought it would be a good >>> idea to start with the existing application first and once it is >>> up and running I can start modifying it. But it just doesn't >>> work for me. >>> >>> Maybe I need to start from scratch again and re-install KIM, >>> Java and Tomcat... >>> >>> From the documentation everything seemed fairly simple but now >>> my head is full of question-marks. >>> >>> Do you have any great ideas? >>> >>> Cheers, >>> >>> Juha >>> >>> P.S. my last attempt resulted in populator tool disappear. It >>> simply doesn't start and I am trying now by re-starting my lap-top. >>> >> >> >> > -- Philip Alexiev Software Engineer Ontotext AD -------------- next part -------------- An HTML attachment was scrubbed... URL: From philip.alexiev at ontotext.com Wed Jun 16 07:34:29 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Wed, 16 Jun 2010 14:34:29 +0300 Subject: [Kim-discussion] Update In-Reply-To: <1783F19CFAA5694C83FB8D718346C5BC07561163@S-DC-EXM12.net1.cec.eu.int> References: <1783F19CFAA5694C83FB8D718346C5BC0756115D@S-DC-EXM12.net1.cec.eu.int> <4C178C31.4020403@ontotext.com> <1783F19CFAA5694C83FB8D718346C5BC0756115E@S-DC-EXM12.net1.cec.eu.int> <54A59B0C-A0AE-434D-99C0-DE6E5BBC2BCE@ontotext.com> <4C187F44.1030105@ontotext.com> <1783F19CFAA5694C83FB8D718346C5BC07561163@S-DC-EXM12.net1.cec.eu.int> Message-ID: <4C18B6C5.6090200@ontotext.com> Hello Juha, Please follow the actions in my previous mail. It is safe to delete all the contents of both $TOMCAT/temp/ and $TOMCAT/work/ folders. You can even delete $TOMCAT/webapps/KIM/ as well and tomcat will unpack it from the web archive again. Actually running tomcat from within KIM is not required. You can easily just deploy the web archive KIM.war in $TOMCAT/webapps/ and start Tomcat the normal way. In the new version of KIM we don't even have a script to start Tomcat. Hope this helps, Philip On 06/16/2010 11:25 AM, Juha.Junttila at ec.europa.eu wrote: > Actually I did use http://localhost:8080/KIM/ - I just forgot to type > the full address in my previous e-mail. I have deployed KIM.war to > $TOMCAT/webapps/ > $TOMCAT/work/ has a number of sub-folders but no files in the > $TOMCAT/work/ -folder - should I delete all the sub-folders? > $TOMCAT/temp/ only has only one file - which is empty: > safeToDelete.tmp - do you think that might be the problem? -- Philip Alexiev Software Engineer Ontotext AD From wolfgang.yang at gmail.com Fri Jun 18 15:29:52 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Fri, 18 Jun 2010 14:29:52 -0500 Subject: [Kim-discussion] question about world kb format Message-ID: Dear List, It seems that the syntax of the KIM world KB is not strictly RDF. Can I use RDF statement in KIM World KB? Thank you! Fangkai From borislav.popov at ontotext.com Sat Jun 19 00:26:17 2010 From: borislav.popov at ontotext.com (borislav popov) Date: Sat, 19 Jun 2010 07:26:17 +0300 Subject: [Kim-discussion] question about world kb format In-Reply-To: References: Message-ID: <8A901416-4C88-482C-BB6C-038F875FC16F@ontotext.com> Yes Yang - you can go ahead and ask if you have issues b On Jun 18, 2010, at 10:29 PM, Yang Fangkai wrote: > Dear List, > > It seems that the syntax of the KIM world KB is not strictly RDF. > Can I use RDF statement in KIM World KB? Thank you! > > Fangkai > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion From wolfgang.yang at gmail.com Mon Jun 21 16:12:50 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Mon, 21 Jun 2010 15:12:50 -0500 Subject: [Kim-discussion] insert new instances into KIM Message-ID: Dear List, I have a questinon about inserting new instances into KIM world KB. I have a RDF file, and use toolRDFImport.cmd to import it. All the entities can be successfully imported. But my question is how to establish a connection between synonyms. For example, I import triples describing a company which locates in "Tokyo". In the imported RDF file, this Tokyo has its unique URI which is different from that in wkb.nt. Is there anyway to automatically align these synonyms? Thank you! Fangkai From wolfgang.yang at gmail.com Mon Jun 21 16:44:51 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Mon, 21 Jun 2010 15:44:51 -0500 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: Message-ID: One follow-up question: I import an entity called "formic acid" which is an instance of class "acid" (class acid is defined to be a subclass of protont:object, and declared visible. The instances is declared as generated by a instance of protons:Trusted), and the RDF file is imported by toolRDFImport.cmd. I do the following queries by WebGUI: 1. Query all entities which are instances of acid, I can see instance "formic acid". 2. Query all entities which are instances of acid and begin with string "formic", it retrieves nothing. 3. Query all entities which are instances of protont#entity, I can see instance "formic acid". 4. Query all entities which are instances of protont#entity and begin with string "formic", it retrieves nothing. I am wondering why this happens. Is there any different treatment between the triples in wkb.nt and the triples imported by toolRDFImport.cmd? Thank you! Fangkai On Mon, Jun 21, 2010 at 3:12 PM, Yang Fangkai wrote: > Dear List, > > ? ? ? I have a questinon about inserting new instances into KIM world > KB. I have a RDF file, and use toolRDFImport.cmd to import it. All the > entities can be successfully imported. But my question is how to > establish a connection between synonyms. For example, I import triples > describing a company which locates in "Tokyo". In the imported RDF > file, this Tokyo has its unique URI which is different from that in > wkb.nt. Is there anyway to automatically align these synonyms? > > Thank you! > > Fangkai > -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu From philip.alexiev at sirma.bg Mon Jun 21 18:13:15 2010 From: philip.alexiev at sirma.bg (Philip Alexiev) Date: Tue, 22 Jun 2010 01:13:15 +0300 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: Message-ID: <4C1FE3FB.4030200@sirma.bg> Hello Yang Could you send me the RDF describing this entity ? It is interesting why you experience such behavior. Thanks, Philip On 06/21/2010 11:44 PM, Yang Fangkai wrote: > One follow-up question: > > I import an entity called "formic acid" which is an instance of class > "acid" (class acid is defined to be a subclass of protont:object, and > declared visible. The instances is declared as generated by a instance > of protons:Trusted), and the RDF file is imported by > toolRDFImport.cmd. > > I do the following queries by WebGUI: > > 1. Query all entities which are instances of acid, I can see instance > "formic acid". > 2. Query all entities which are instances of acid and begin with > string "formic", it retrieves nothing. > 3. Query all entities which are instances of protont#entity, I can see > instance "formic acid". > 4. Query all entities which are instances of protont#entity and begin > with string "formic", it retrieves nothing. > > I am wondering why this happens. Is there any different treatment > between the triples in wkb.nt and the triples imported by > toolRDFImport.cmd? > > Thank you! > > Fangkai > From wolfgang.yang at gmail.com Mon Jun 21 18:45:55 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Mon, 21 Jun 2010 17:45:55 -0500 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: <4C1FE3FB.4030200@sirma.bg> References: <4C1FE3FB.4030200@sirma.bg> Message-ID: Thank you, Philip, Currently I am out of my office, so I will send you the rdf file tomorrow. Do you think I must declare each label of the entity as a "string" datatype before using the string-matching query mode such as "startwith"? Fangkai On Mon, Jun 21, 2010 at 5:13 PM, Philip Alexiev wrote: > Hello Yang > > Could you send me the RDF describing this entity ? It is interesting why you > experience such behavior. > > Thanks, > Philip > > > On 06/21/2010 11:44 PM, Yang Fangkai wrote: >> >> One follow-up question: >> >> I import an entity called "formic acid" which is an instance of class >> "acid" (class acid is defined to be a subclass of protont:object, and >> declared visible. The instances is declared as generated by a instance >> of protons:Trusted), and the RDF file is imported by >> toolRDFImport.cmd. >> >> I do the following queries by WebGUI: >> >> 1. Query all entities which are instances of acid, I can see instance >> "formic acid". >> 2. Query all entities which are instances of acid and begin with >> string "formic", it retrieves nothing. >> 3. Query all entities which are instances of protont#entity, I can see >> instance "formic acid". >> 4. Query all entities which are instances of protont#entity and begin >> with string "formic", it retrieves nothing. >> >> I am wondering why this happens. Is there any different treatment >> between the triples in wkb.nt and the triples imported by >> toolRDFImport.cmd? >> >> Thank you! >> >> Fangkai >> > From philip.alexiev at sirma.bg Tue Jun 22 03:26:55 2010 From: philip.alexiev at sirma.bg (Philip Alexiev) Date: Tue, 22 Jun 2010 10:26:55 +0300 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: <4C1FE3FB.4030200@sirma.bg> Message-ID: <4C2065BF.5040908@sirma.bg> No , the entity having a label should be sufficient: "label" . On 06/22/2010 01:45 AM, Yang Fangkai wrote: > Currently I am out of my office, so I will send you the rdf > file tomorrow. Do you think I must declare each label of the entity as > a "string" datatype before using the string-matching query mode such > as "startwith"? > From wolfgang.yang at gmail.com Tue Jun 22 09:27:40 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Tue, 22 Jun 2010 08:27:40 -0500 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: <4C2065BF.5040908@sirma.bg> References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> Message-ID: Hello, Philip, Please find the attached file. In this file, there are three "drill_fluid_additive" instances whose names begin with "SAFE". All classes and properties have been defined in protont,owl properly (and with a rdfs:label attribute for each class, in particular, drill_fluid_additive is defined as a subclass of protont#material which is a subclass of protont#object), and declared visible. After importing this file by toolRDFImport.cmd, I have the following problem: 1. query: query instances for all "drill_fluid_additive", all instances are retrieved; query instances for all "drill_fluid_additive" starting with "SAFE", no result is returned. The same problem occurs when I use these triples (either in RDF or in N-Triple format) as a separate knowledge base that loaded by Sesame, instead of importing triples by toolRDFImport. 2. In Protege, I define a restriction "producedBy" for drill_fluid_additive, linking the entity to its manufacturers, but in WebGUI, after selecting X as an entity of "drill_fluid_additive", I cannot select property "producedBy" in the drag-down box below it. But for other three properties (e.g., primaryFunction, etc.) I can. I am wondering why. 3. The manufacturer locates in Houston, TX. How can I relate the "Houston" in this newly imported RDF to the city "Houston" defined in KIM wkb.nt? Because I want to take advantage of the underlying geographic information in KIM. Thank you so much for your help! Fangkai On Tue, Jun 22, 2010 at 2:26 AM, Philip Alexiev wrote: > No , the entity having a label should be sufficient: > > "label" . > > > On 06/22/2010 01:45 AM, Yang Fangkai wrote: >> >> ? ? ? ? ? Currently I am out of my office, so I will send you the rdf >> file tomorrow. Do you think I must declare each label of the entity as >> a "string" datatype before using the string-matching query mode such >> as "startwith"? >> > > -------------- next part -------------- A non-text attachment was scrubbed... Name: test.rdf Type: text/xml Size: 5437 bytes Desc: not available URL: From philip.alexiev at sirma.bg Tue Jun 22 11:03:30 2010 From: philip.alexiev at sirma.bg (Philip Alexiev) Date: Tue, 22 Jun 2010 18:03:30 +0300 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> Message-ID: <4C20D0C2.9040208@sirma.bg> Hello Yang, You will find the answers inline the document. On 06/22/2010 04:27 PM, Yang Fangkai wrote: > Hello, Philip, > > Please find the attached file. In this file, there are > three "drill_fluid_additive" instances whose names begin with "SAFE". > All classes and properties have been defined in protont,owl properly > (and with a rdfs:label attribute for each class, in particular, > drill_fluid_additive is defined as a subclass of protont#material > which is a subclass of protont#object), and declared visible. After > importing this file by toolRDFImport.cmd, I have the following > problem: > > 1. query: query instances for all "drill_fluid_additive", > all instances are retrieved; query instances for all > "drill_fluid_additive" starting with "SAFE", no result is returned. > The same problem occurs when I use these triples (either in RDF or in > N-Triple format) as a separate knowledge base that loaded by Sesame, > instead of importing triples by toolRDFImport. > Basically, KIM operates in two modes - using labels, or aliases. This is configured in $KIM_HOME/config/install.properties in com.ontotext.kim.KIMConstants.ENTITY_DESCR property. - Labels - more simple and more compact. Generally what is required is that each entity has at least a protons:mainLabel, and different optional rdfs:label descriptions. - Aliases - uses aliases, which are objects by themselves, to describe the entities. The idea is to be able to have meta information for the labels. This is the default model KIM uses. In your case you provided a label, but your model is most probably "Aliases". And, of course, you should also provide a mainAlias/mainLabel . > 2. In Protege, I define a restriction "producedBy" for > drill_fluid_additive, linking the entity to its manufacturers, but in > WebGUI, after selecting X as an entity of "drill_fluid_additive", I > cannot select property "producedBy" in the drag-down box below it. But > for other three properties (e.g., primaryFunction, etc.) I can. I am > wondering why. > Try setting "drill_fluid_additive" as a domain of the property "producedBy" . > 3. The manufacturer locates in Houston, TX. How can I relate > the "Houston" in this newly imported RDF to the city "Houston" defined > in KIM wkb.nt? Because I want to take advantage of the underlying > geographic information in KIM. > Use the URI of Houston from the wkb.nt : . > Thank you so much for your help! > > Fangkai > Hope this helps, Philip From wolfgang.yang at gmail.com Tue Jun 22 11:12:46 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Tue, 22 Jun 2010 10:12:46 -0500 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: <4C20D0C2.9040208@sirma.bg> References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> <4C20D0C2.9040208@sirma.bg> Message-ID: Thank you, Philip. I will try what you suggest and let you know if I have more questions. One more thing is: I have imported the triples already. After I add the mainAlias/mainLabel to the RDF file, shall I simply import all of them again? It seems that when I import a file twice, there will be two entities in the knowledge base. Is there anyway to clear the old-version imported triples? Fangkai On Tue, Jun 22, 2010 at 10:03 AM, Philip Alexiev wrote: > Hello Yang, > > You will find the answers inline the document. > > > On 06/22/2010 04:27 PM, Yang Fangkai wrote: >> >> Hello, Philip, >> >> ? ? ? ? ? ? Please find the attached file. In this file, there are >> three "drill_fluid_additive" instances whose names begin with "SAFE". >> All classes and properties have been defined in protont,owl properly >> (and with a rdfs:label attribute for each class, in particular, >> drill_fluid_additive is defined as a subclass of protont#material >> which is a subclass of protont#object), and declared visible. After >> importing this file by toolRDFImport.cmd, I have the following >> problem: >> >> ? ? ? ? ? 1. query: query instances for all "drill_fluid_additive", >> all instances are retrieved; query instances for all >> "drill_fluid_additive" starting with "SAFE", no result is returned. >> The same problem occurs when I use these triples (either in RDF or in >> N-Triple format) as a separate knowledge base that loaded by Sesame, >> instead of importing triples by toolRDFImport. >> > > Basically, KIM operates in two modes - using labels, or aliases. This is > configured in ?$KIM_HOME/config/install.properties in > ?com.ontotext.kim.KIMConstants.ENTITY_DESCR property. > - Labels - more simple and more compact. Generally what is required is that > each entity has at least a protons:mainLabel, and different optional > rdfs:label descriptions. > - Aliases - uses aliases, which are objects by themselves, to describe the > entities. The idea is to be able to have meta information for the labels. > This is the default model KIM uses. > In your case you provided a label, but your model is most probably > "Aliases". And, of course, you should also provide a mainAlias/mainLabel . >> >> ? ? ? ? ?2. In Protege, I define a restriction "producedBy" for >> drill_fluid_additive, linking the entity to its manufacturers, but in >> WebGUI, after selecting X as an entity of "drill_fluid_additive", I >> cannot select property "producedBy" in the drag-down box below it. But >> for other three properties (e.g., primaryFunction, etc.) I can. I am >> wondering why. >> > > Try setting "drill_fluid_additive" as a domain of the property "producedBy" > . >> >> ? ? ? ? 3. The manufacturer locates in Houston, TX. How can I relate >> the "Houston" in this newly imported RDF to the city "Houston" defined >> in KIM wkb.nt? Because I want to take advantage of the underlying >> geographic information in KIM. >> > > Use the URI of Houston from the wkb.nt : > > . >> >> Thank you so much for your help! >> >> Fangkai >> > > > Hope this helps, > Philip > -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu From philip.alexiev at sirma.bg Tue Jun 22 11:27:36 2010 From: philip.alexiev at sirma.bg (Philip Alexiev) Date: Tue, 22 Jun 2010 18:27:36 +0300 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> <4C20D0C2.9040208@sirma.bg> Message-ID: <4C20D668.1030209@sirma.bg> The first time KIM starts it reads the rdf from the files described in the "imports" section in Sesame.conf . Each subsequent run works with the cache and does not read the rdfs. So you will have to either clear the cache , by removing $KIM_HOME/context/default/populated , include your custom rdf in the imports of Sesame and start KIM again. Or import them in an already running KIM with toolRdfImport. Importing the same statement more than once will result in a single statement in the semantic repository. Cheers, Philip On 06/22/2010 06:12 PM, Yang Fangkai wrote: > One more thing is: I have imported the triples already. After I add > the mainAlias/mainLabel to the RDF file, shall I simply import all of > them again? It seems that when I import a file twice, there will be > two entities in the knowledge base. Is there anyway to clear the > old-version imported triples? > > From wolfgang.yang at gmail.com Tue Jun 22 14:07:20 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Tue, 22 Jun 2010 13:07:20 -0500 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: <4C20D0C2.9040208@sirma.bg> References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> <4C20D0C2.9040208@sirma.bg> Message-ID: Hello, Philip, Question 1 and 3 have been solved, but it seems question 2 is still unresolved. In Protege, I have added Drill_fluid_additive as a domain of producedBy, but in WebGUI it still doesn't show up... Thank you so much for your help! Fangkai On Tue, Jun 22, 2010 at 10:03 AM, Philip Alexiev wrote: > Hello Yang, > > You will find the answers inline the document. > > > On 06/22/2010 04:27 PM, Yang Fangkai wrote: >> >> Hello, Philip, >> >> ? ? ? ? ? ? Please find the attached file. In this file, there are >> three "drill_fluid_additive" instances whose names begin with "SAFE". >> All classes and properties have been defined in protont,owl properly >> (and with a rdfs:label attribute for each class, in particular, >> drill_fluid_additive is defined as a subclass of protont#material >> which is a subclass of protont#object), and declared visible. After >> importing this file by toolRDFImport.cmd, I have the following >> problem: >> >> ? ? ? ? ? 1. query: query instances for all "drill_fluid_additive", >> all instances are retrieved; query instances for all >> "drill_fluid_additive" starting with "SAFE", no result is returned. >> The same problem occurs when I use these triples (either in RDF or in >> N-Triple format) as a separate knowledge base that loaded by Sesame, >> instead of importing triples by toolRDFImport. >> > > Basically, KIM operates in two modes - using labels, or aliases. This is > configured in ?$KIM_HOME/config/install.properties in > ?com.ontotext.kim.KIMConstants.ENTITY_DESCR property. > - Labels - more simple and more compact. Generally what is required is that > each entity has at least a protons:mainLabel, and different optional > rdfs:label descriptions. > - Aliases - uses aliases, which are objects by themselves, to describe the > entities. The idea is to be able to have meta information for the labels. > This is the default model KIM uses. > In your case you provided a label, but your model is most probably > "Aliases". And, of course, you should also provide a mainAlias/mainLabel . >> >> ? ? ? ? ?2. In Protege, I define a restriction "producedBy" for >> drill_fluid_additive, linking the entity to its manufacturers, but in >> WebGUI, after selecting X as an entity of "drill_fluid_additive", I >> cannot select property "producedBy" in the drag-down box below it. But >> for other three properties (e.g., primaryFunction, etc.) I can. I am >> wondering why. >> > > Try setting "drill_fluid_additive" as a domain of the property "producedBy" > . >> >> ? ? ? ? 3. The manufacturer locates in Houston, TX. How can I relate >> the "Houston" in this newly imported RDF to the city "Houston" defined >> in KIM wkb.nt? Because I want to take advantage of the underlying >> geographic information in KIM. >> > > Use the URI of Houston from the wkb.nt : > > . >> >> Thank you so much for your help! >> >> Fangkai >> > > > Hope this helps, > Philip > -- Fangkai Yang, Ph.D student Taylor Hall 3.150A Department of Computer Sciences The University of Texas at Austin Austin, 78712-0233, Texas USA http://www.cs.utexas.edu/~fkyang email: fkyang at cs.utexas.edu From wolfgang.yang at gmail.com Tue Jun 22 17:06:19 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Tue, 22 Jun 2010 16:06:19 -0500 Subject: [Kim-discussion] annotate pdf file Message-ID: Dear List, It seems that the default settings of KIM doesn't support annotating PDF files? I am using KIM2.4 and toolPopulate GUI over a folder containing a bunch of pdf files with suffix .pdf, but it doesn't annotate. Is there any link that tells how to configure KIM to do this annotation? Thank you! Fangkai From philip.alexiev at ontotext.com Wed Jun 23 03:28:38 2010 From: philip.alexiev at ontotext.com (Philip Alexiev) Date: Wed, 23 Jun 2010 10:28:38 +0300 Subject: [Kim-discussion] annotate pdf file In-Reply-To: References: Message-ID: <4C21B7A6.5040206@ontotext.com> Helo Fangkai You can inspect your populater configuration file - $KIM_HOME/config/populater.xml . The option "INPUT_DOC_EXT" holds a list of extensions KIM supports. Add "pdf" there and PDFs will be annotated. We have changed it to be the default in KIM 3.0 . Greetings, Philip On 06/23/2010 12:06 AM, Yang Fangkai wrote: > Dear List, > > It seems that the default settings of KIM doesn't support > annotating PDF files? I am using KIM2.4 and toolPopulate GUI over a > folder containing a bunch of pdf files with suffix .pdf, but it > doesn't annotate. Is there any link that tells how to configure KIM to > do this annotation? > > Thank you! > > Fangkai > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion > > -- Philip Alexiev Software Engineer Ontotext AD From philip.alexiev at sirma.bg Wed Jun 23 03:35:48 2010 From: philip.alexiev at sirma.bg (Philip Alexiev) Date: Wed, 23 Jun 2010 10:35:48 +0300 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> <4C20D0C2.9040208@sirma.bg> Message-ID: <4C21B954.4060701@sirma.bg> Try to remove Tomcat's cache ( $TOMCAT/webapps/KIM/ , $TOMCAT/work/* ) Also make sure, that the class "drill_fluid_additive" is visible, that is - this statement exists in the semantic repository: "" . On 06/22/2010 09:07 PM, Yang Fangkai wrote: > Hello, Philip, > > Question 1 and 3 have been solved, but it seems question 2 is > still unresolved. In Protege, I have added Drill_fluid_additive as a > domain of producedBy, but in WebGUI it still doesn't show up... > > Thank you so much for your help! > > Fangkai > > On Tue, Jun 22, 2010 at 10:03 AM, Philip Alexiev > wrote: > >> Hello Yang, >> >> You will find the answers inline the document. >> >> >> On 06/22/2010 04:27 PM, Yang Fangkai wrote: >> >>> Hello, Philip, >>> >>> Please find the attached file. In this file, there are >>> three "drill_fluid_additive" instances whose names begin with "SAFE". >>> All classes and properties have been defined in protont,owl properly >>> (and with a rdfs:label attribute for each class, in particular, >>> drill_fluid_additive is defined as a subclass of protont#material >>> which is a subclass of protont#object), and declared visible. After >>> importing this file by toolRDFImport.cmd, I have the following >>> problem: >>> >>> 1. query: query instances for all "drill_fluid_additive", >>> all instances are retrieved; query instances for all >>> "drill_fluid_additive" starting with "SAFE", no result is returned. >>> The same problem occurs when I use these triples (either in RDF or in >>> N-Triple format) as a separate knowledge base that loaded by Sesame, >>> instead of importing triples by toolRDFImport. >>> >>> >> Basically, KIM operates in two modes - using labels, or aliases. This is >> configured in $KIM_HOME/config/install.properties in >> com.ontotext.kim.KIMConstants.ENTITY_DESCR property. >> - Labels - more simple and more compact. Generally what is required is that >> each entity has at least a protons:mainLabel, and different optional >> rdfs:label descriptions. >> - Aliases - uses aliases, which are objects by themselves, to describe the >> entities. The idea is to be able to have meta information for the labels. >> This is the default model KIM uses. >> In your case you provided a label, but your model is most probably >> "Aliases". And, of course, you should also provide a mainAlias/mainLabel . >> >>> 2. In Protege, I define a restriction "producedBy" for >>> drill_fluid_additive, linking the entity to its manufacturers, but in >>> WebGUI, after selecting X as an entity of "drill_fluid_additive", I >>> cannot select property "producedBy" in the drag-down box below it. But >>> for other three properties (e.g., primaryFunction, etc.) I can. I am >>> wondering why. >>> >>> >> Try setting "drill_fluid_additive" as a domain of the property "producedBy" >> . >> >>> 3. The manufacturer locates in Houston, TX. How can I relate >>> the "Houston" in this newly imported RDF to the city "Houston" defined >>> in KIM wkb.nt? Because I want to take advantage of the underlying >>> geographic information in KIM. >>> >>> >> Use the URI of Houston from the wkb.nt : >> >> . >> >>> Thank you so much for your help! >>> >>> Fangkai >>> >>> >> >> Hope this helps, >> Philip >> >> > > > From wolfgang.yang at gmail.com Wed Jun 23 11:00:53 2010 From: wolfgang.yang at gmail.com (Yang Fangkai) Date: Wed, 23 Jun 2010 10:00:53 -0500 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: <4C21B954.4060701@sirma.bg> References: <4C1FE3FB.4030200@sirma.bg> <4C2065BF.5040908@sirma.bg> <4C20D0C2.9040208@sirma.bg> <4C21B954.4060701@sirma.bg> Message-ID: Hello, Philip, It still doesn't work... Currently the domain of producedBy is protonu#ProductModel and protont#Drill_Fluid_Additive, and the range is protont#agent and protont#FluidSupplier. I am wondering if the definition is problematic? Sorry for bothering you so much and thank you for your suggestion! Fangkai On Wed, Jun 23, 2010 at 2:35 AM, Philip Alexiev wrote: > > Try to remove Tomcat's cache ?( $TOMCAT/webapps/KIM/ , $TOMCAT/work/* ) > > Also make sure, that the class "drill_fluid_additive" is visible, that is - > this statement exists in the semantic repository: > > > "" ?. > > > On 06/22/2010 09:07 PM, Yang Fangkai wrote: >> >> Hello, Philip, >> >> ? ? ? ? Question 1 and 3 have been solved, but it seems question 2 is >> still unresolved. In Protege, I have added Drill_fluid_additive as a >> domain of producedBy, but in WebGUI it still doesn't show up... >> >> Thank you so much for your help! >> >> Fangkai >> >> On Tue, Jun 22, 2010 at 10:03 AM, Philip Alexiev >> ?wrote: >> >>> >>> Hello Yang, >>> >>> You will find the answers inline the document. >>> >>> >>> On 06/22/2010 04:27 PM, Yang Fangkai wrote: >>> >>>> >>>> Hello, Philip, >>>> >>>> ? ? ? ? ? ? Please find the attached file. In this file, there are >>>> three "drill_fluid_additive" instances whose names begin with "SAFE". >>>> All classes and properties have been defined in protont,owl properly >>>> (and with a rdfs:label attribute for each class, in particular, >>>> drill_fluid_additive is defined as a subclass of protont#material >>>> which is a subclass of protont#object), and declared visible. After >>>> importing this file by toolRDFImport.cmd, I have the following >>>> problem: >>>> >>>> ? ? ? ? ? 1. query: query instances for all "drill_fluid_additive", >>>> all instances are retrieved; query instances for all >>>> "drill_fluid_additive" starting with "SAFE", no result is returned. >>>> The same problem occurs when I use these triples (either in RDF or in >>>> N-Triple format) as a separate knowledge base that loaded by Sesame, >>>> instead of importing triples by toolRDFImport. >>>> >>>> >>> >>> Basically, KIM operates in two modes - using labels, or aliases. This is >>> configured in ?$KIM_HOME/config/install.properties in >>> ?com.ontotext.kim.KIMConstants.ENTITY_DESCR property. >>> - Labels - more simple and more compact. Generally what is required is >>> that >>> each entity has at least a protons:mainLabel, and different optional >>> rdfs:label descriptions. >>> - Aliases - uses aliases, which are objects by themselves, to describe >>> the >>> entities. The idea is to be able to have meta information for the labels. >>> This is the default model KIM uses. >>> In your case you provided a label, but your model is most probably >>> "Aliases". And, of course, you should also provide a mainAlias/mainLabel >>> . >>> >>>> >>>> ? ? ? ? ?2. In Protege, I define a restriction "producedBy" for >>>> drill_fluid_additive, linking the entity to its manufacturers, but in >>>> WebGUI, after selecting X as an entity of "drill_fluid_additive", I >>>> cannot select property "producedBy" in the drag-down box below it. But >>>> for other three properties (e.g., primaryFunction, etc.) I can. I am >>>> wondering why. >>>> >>>> >>> >>> Try setting "drill_fluid_additive" as a domain of the property >>> "producedBy" >>> . >>> >>>> >>>> ? ? ? ? 3. The manufacturer locates in Houston, TX. How can I relate >>>> the "Houston" in this newly imported RDF to the city "Houston" defined >>>> in KIM wkb.nt? Because I want to take advantage of the underlying >>>> geographic information in KIM. >>>> >>>> >>> >>> Use the URI of Houston from the wkb.nt : >>> ? >>> ?. >>> >>>> >>>> Thank you so much for your help! >>>> >>>> Fangkai >>>> >>>> >>> >>> Hope this helps, >>> Philip >>> >>> >> >> >> > > From mnozchev at sirma.bg Fri Jun 25 04:00:39 2010 From: mnozchev at sirma.bg (Marin Nozhchev) Date: Fri, 25 Jun 2010 11:00:39 +0300 Subject: [Kim-discussion] insert new instances into KIM In-Reply-To: References: Message-ID: <4C246227.6030204@sirma.bg> Dear Fangkai, I can give you a detailed answer of the particular question as it tackles an important issue . I hope it's not too late. In RDF, you can have multiple labels (rdfs:label) for a single entity like Tokyo, but Tokyo must have a single URI. However, it's impossible to make sure that all RDF vocabularies published in the world use the same URIs. So you have three options: 1. Preprocess the smaller RDF file, replacing the different URIs of an entity with a single one. This will require the most initial effort but there will be no downsides afterwards. 2. Add owl:sameAs statements that link the equivalent entities, eg. wkb:Tokyo owl:sameAs other:Tokyo . This will make them equivalent at database level - that is - the RDF database behind KIM will merge the data (e.g. companies located in) about both URIs together. The database will also return both URIs as result to queries that match the data about any one of them. In practice, however, this leads to both ambiguity and increased load on the database. The theoretical behavior is undefined in some corner cases and is generally difficult to implement by the database. KIM 3 with BigOWLIM 3.3 as RDF database has improved support for owl:sameAs. We haven't tested sameAs support in KIM 2.4 with SwiftOWLIM 2.0 DB, so I can't promise anything there. Learn more about sameAs here http://journal.webscience.org/403/2/websci10_submission_123.pdf 3. Add semantic metadata in the RDF that explains that the two concepts can be used interchangeably. The SKOS mapping property skos:exactMatch is the best candidate e.g. wkb:Tokyo skos:exactMatch other:Tokyo (full URI of skos:exactMatch : http://www.w3.org/2004/02/skos/core#exactMatch) . skos:exactMatch doesn't mean anything to the RDF database - the triples will not be interpreted differently than any other data that refers to the two Tokyo URIs. In this case, the client application - KIM or your custom client - should take this metadata into account. KIM 2.4 doesn't support SKOS and KIM 3.0 (available now!) has limited support for SKOS semantics. Technically, you can alternatively add schema statements and inference rules to implement skos:exactMatch semantic directly in the database, but in that case you are better off just using owl:sameAs as the same problems in corner cases will pop up again. Learn more about SKOS here - http://www.w3.org/TR/2009/NOTE-skos-primer-20090818/ and see the difference between owl:sameAs and skos:exactMatch here - http://www.w3.org/TR/2009/REC-skos-reference-20090818/#L4858 . As you can see, only option 1 is a complete solution. Using the other options require that you both upgrade KIM and read the reference material so you know what to expect. Best regards, Marin Nozhchev On 21.06.2010 23:12 ?., Yang Fangkai wrote: > Dear List, > > I have a questinon about inserting new instances into KIM world > KB. I have a RDF file, and use toolRDFImport.cmd to import it. All the > entities can be successfully imported. But my question is how to > establish a connection between synonyms. For example, I import triples > describing a company which locates in "Tokyo". In the imported RDF > file, this Tokyo has its unique URI which is different from that in > wkb.nt. Is there anyway to automatically align these synonyms? > > Thank you! > > Fangkai > _______________________________________________ > Kim-discussion mailing list > Kim-discussion at ontotext.com > http://ontotext.com/mailman/listinfo/kim-discussion >